Po-Wen Shih
We explore methods for efficiently walking through the interior of passageways such as galleries, arteries, ducts, tunnels, pipes, corridors, and caves. Common to the interior of all those passageways is that only a small portion of the labyrinth is visible at anytime. We pre-process the model and collect information that assists us to render, at any point, only a small subset of the model. First, the model is embedded in a regular grid of cells. In a pre-processing stage, a novel algorithm collects, for each cell, the list of all potentially visible cells. The list is then region-partitioned and depth-sorted to further improve performance by supporting view-cone and depth queuing. We propose extensions that will improve pre-processing time by adaptive grid refinement and decrease memory requirement by incremental storage.
For more information, please see my personal homepage.
Last modified: October 6, 1996
Torsten Möller / moeller@cis.ohio-state.edu